Carpenter 2008.

Methods	Design: randomised controlled trial Follow‐up: 24 weeks post‐treatment only Setting: two university‐affiliated, outpatient community‐based treatment programs Country: USA
Participants	Participants: 38 methadone‐maintained DSM‐IV opiate‐dependent patients with major depressive disorder/dysthymia on the Structured Clinical Interview‐Substance Abuse Comorbidity (SCID‐SAC) Mean age (years): BTDD 38.8 (SD = 10.4); REL 41.2 (SD = 10.9) Sex: 58% male Ethnicity: 82% Caucasian
Interventions	Description: Behavioral Therapy for Depression in Drug Dependence (BTDD) versus Structured Relaxation Intervention (REL). BTDD incorporated aspects of three operant‐based treatment programs: Changing Reinforcing Events (Lewinsohn 1980), the Community Reinforcement Approach (Meyers 1995), and Treatment Plan Contingency Management (points were earned for participating in the therapy session (3 points) and completing out‐of‐session assignments (up to 10 points); maximum of 208 points (1 point = 1 dollar) could be earned for 100% attendance (72 points) and completion of out‐of‐session assignments (136 points). Points were exchangeable for goods and services consistent with the treatment plan (Iguchi 1997). REL was based on a structured manual (Brown 1997) focusing on three relaxation strategies: 1) progressive muscle relaxation, 2) autogenic relaxation exercises, and 3) visual imagery based on idiographic scenarios of relaxation or tranquility. Format: group therapy; plus community‐based methadone treatment programs Duration: 24 weekly sessions Allocation: BTDD n = 18 ; REL n = 20
Outcomes	Treatment attendance: mean number of sessions attended Self‐report: BDI mean (SD) total score (past 7 days) Interviewer‐rated: HDRS (SD) mean total score (past 7 days); Substance Use Weekly Inventory (SUI modification of TLFB) ‐ percentage of weeks cocaine, opiate and benzodiazepine use during treatment (past 30 days; data collected weekly for 24 weeks). Attrition: proportion of participants with missing 24 weeks post‐treatment data
Notes	Fidelity: all therapists completed a BTDD or Relaxation Therapy Checklist following each treatment session. Checklists outlined key components of each respective treatment and provided a means to assess adherence to each therapy condition. Session audiotapes were reviewed by senior therapists to monitor adherence to the treatment structure. Funding: NIDA grants R01 DA13118, K02 DA00288, K24 DA021850 (Nunes), K23 DA021850 (Carpenter). Conflict of interest: no statement provided.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote (via author email): "The randomisation scheme (blocks) were generated by an independent person (based on a random number process)" Quote: "The treatment groups did not differ significantly on most baseline measures. Although stratification procedures balanced the proportion of participants using illicit substances in the week before treatment between the two treatment groups, the proportion of days that opiates were used in the month prior to treatment was greater among those in BTDD. (p. 647)"
Allocation concealment (selection bias)	Low risk	Quote (via author email): "The randomisation grid was maintained in a private password account held by the independent person." Comment: investigators enrolling participants could not foresee assignment.
Blinding of participants and personnel (performance bias) subjective outcomes	High risk	Participants and therapists were not blinded to treatment allocation.
Blinding of participants and personnel (performance bias) objective outcomes	High risk	Participant and personnel knowledge of group allocation may influence treatment engagement/participation/ attendance.
Blinding of outcome assessment (detection bias) self‐report	Unclear risk	Quote (vie author email): "The self‐report assessments were to be given by the research assistant ‐ who may or may not have been blind to therapy condition." Comment: The authors judge risk of bias to be unclear. While participants were not blind to treatment allocation and self‐report measures may be impacted by self‐presentation bias or client insight, it is unlikely that any such risk of bias will vary by treatment condition. The research assistant administering the self‐report measures is unlikely to have influenced the results.
Blinding of outcome assessment (detection bias) interviewer‐rated outcomes	High risk	Quote (via author email): "It could not be guaranteed they were blind to condition."
Blinding of outcome assessment (detection bias) treatment retention/attendance	Low risk	Blinding unclear, but outcome measure is unlikely to be influenced by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	High risk	Treatment attendance: BTDD n = 11.6 (SD=6.1); REL n = 15.3 (SD = 6.9); (P = 0.10) Comment: the number of treatment sessions completed was similar and did not differ significantly between groups Outcome assessment: attrition total n = 38, 37%; BTDD n = 9, 50%; REL n = 15, 25%; (P = 0 .05) Quote: "No significant differences on baseline measures between those participants completing the study and those who dropped out. (p. 647)." Comment: BTDD group had double the attrition rate of the REL group; LOCF was used for missing data and ITT was not used for the endpoint analysis.
Selective reporting (reporting bias)	Unclear risk	Risk is unclear as no study protocol is available. The results of all outcomes included in the method section of the paper are reported.
Other bias	Unclear risk	Fidelity: therapist completed checklists for each session. Recordings were reviewed by senior therapists but no independent fidelity ratings were conducted. Other treatments: all participants were receiving at least 60 mg of methadone, it is unclear whether dosage was consistent between groups. Randomisation was stratified by antidepressant medication, and antidepressant medication was controlled in outcome analyses with a null effect reported for depression outcomes. It is unclear whether participants were able to receive other psychotherapy, in addition to that provided through the study.